Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Page created by Arthur Rodgers
 
CONTINUE READING
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
rur.oekom.de      https://doi.org/10.14512/rur.63

BEITRAG          
                     ARTICLE                                                                                           OPEN ACCESS

Rail Accessibility in Germany: Changing Regional
Disparities between 1990 and 2020

Fabian Wenner        , Alain Thierstein

Received: 22 May 2020      Accepted: 18 January 2021      Published online: 26 February 2021

Abstract                                                                  Erreichbarkeit im Schienenverkehr in
Transport accessibility is an important location factor for               Deutschland: Sich wandelnde regionale
households and firms. In the last few decades, technological              Disparitäten zwischen 1990 und 2020
and social developments have contributed to a reinvigorated
role of passenger transport. However, rail accessibility is un-           Zusammenfassung
evenly distributed in space. The introduction of high-speed               Verkehrliche Erreichbarkeit stellt einen wichtigen Standort-
rail has furthermore promoted a polarisation of accessibility             faktor für Haushalte und Unternehmen dar. In den letzten
between metropolises and peripheral areas in some European                Jahrzehnten haben technologische und soziale Entwicklungen
countries. In this paper we analyse the development of rail               zu einer neuen Attraktivität des Schienenpersonenverkehrs
accessibility at the regional level in Germany between 1990               beigetragen. Die Erreichbarkeit über den Schienenverkehr
and 2020 for 266 functional city-regions. Our results show                fällt jedoch räumlich sehr unterschiedlich aus. Die Einfüh-
two different facets: the number of regions that are directly             rung des Hochgeschwindigkeitsverkehrs hat zudem in einigen
connected to one another has decreased, but at the same                   europäischen Ländern eine Polarisierung der Erreichbarkeit
time the spatial disparities of accessibility have decreased,             zwischen Metropolen und peripheren Räumen befördert. In
albeit to a small extent. This development was strongest in               diesem Beitrag analysieren wir die Entwicklung der Bahner-
East Germany after German reunification and thus largely                  reichbarkeit auf regionaler Ebene in Deutschland zwischen
a consequence of the renovation of the conventional rail                  1990 und 2020 für 266 funktionale Stadtregionen. Unsere Er-
infrastructure, not high-speed rail. Nevertheless, it can be              gebnisse zeigen zwei unterschiedliche Facetten: Die Zahl der
concluded that the introduction of high-speed traffic in Ger-             direkt miteinander verbundenen Regionen hat sich verringert,
many did not lead to an increase in accessibility disparities.            aber zugleich zeigt sich für die Erreichbarkeit der Bevölkerung
Instead, the accessibility effects of high-speed rail in Germany          eine Abschwächung der räumlichen Disparitäten, wenn auch
seem to break the traditional dichotomy between core and                  in geringem Maße. Diese Entwicklung war in Ostdeutschland
periphery.                                                                nach der deutschen Wiedervereinigung am stärksten und da-
                                                                          mit weitgehend eine Folge der Sanierung der konventionellen
Keywords: Accessibility · Rail · High Speed Rail · Regional               Schieneninfrastruktur, nicht des Hochgeschwindigkeitsver-
Disparities · Germany                                                     kehrs. Dennoch kann der Schluss gezogen werden, dass seine
                                                                          Einführung in Deutschland nicht zur Erhöhung von Erreich-
                                                                          barkeitsdisparitäten geführt hat. Stattdessen scheinen die
 Fabian Wenner, Lehrstuhl Raumentwicklung, Technische                    Erreichbarkeitswirkungen des Hochgeschwindigkeitsverkehrs
Universität München, Arcisstraße 21, 80333 München
                                                                          in Deutschland die traditionelle Dichotomie zwischen Kern
f.wenner@tum.de
                                                                          und Peripherie zu durchbrechen.
Prof. Dr. Alain Thierstein, Lehrstuhl Raumentwicklung, Technische
Universität München, Arcisstraße 21, 80333 München                        Schlüsselwörter: Erreichbarkeit · Bahn · Hoch-
thierstein@tum.de                                                         geschwindigkeits-Bahnverkehr · regionale Disparitäten ·
                                                                          Deutschland
          © 2021 Wenner; licensee oekom verlag. This Open
Access article is published under the Creative Commons
Attribution-ShareAlike 4.0 International Licence.

Raumforschung und Raumordnung | Spatial Research and Planning            (2021) 0/0: 1–21                                                   1
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
F. Wenner, A. Thierstein

1 Introduction                                                     cent decades have seen an increased profit orientation and
                                                                   the (partial) privatisation of several national rail providers
Accessibility is an important determinant of regional and          throughout Europe with adverse consequences for regional
local development (Clark 1958). Transport infrastructure           accessibility in some countries, as peripheral and tangen-
and services expand the opportunities for households to            tial lines often have the lowest internal profitability. On the
access distant jobs and amenities, and support, inter alia,        other hand, progressing European integration has in some
the emergence of agglomeration benefits for firms. Hence,            cases reduced the peripherality of border regions in the rail
accessible places and regions have long attracted economic         network.
activity (Axhausen 2008: 5).                                           Despite numerous (mostly ex-ante) studies in several Eu-
    After a phase of relative decline, passenger rail transport    ropean countries and for Europe as a whole, there have been
has experienced a resurgence as a transport mode in Eu-            few assessments of the development of rail accessibility in
rope during recent decades. Several societal and technolog-        the case of Germany. During the last three decades, the
ical developments have contributed to this reinvigoration.         rail network has been influenced by the triple effects of re-
Tertiarisation and the rise of the knowledge economy have          unification, the de jure privatisation of the railway service
increased demand for face-to-face communication and busi-          provider, and the construction of more than 1,000 km of
ness travel (Hall/Pain 2006: 7; Thierstein/Lüthi/Kruse et al.      new high-speed rail infrastructure, which has reshuffled re-
2008). This has supported reurbanisation and a tendency            gional accessibility. Accessibility studies in countries with
towards polycentric development which involves the major           a monocentric urban structure and radial network develop-
metropolitan cores (Münter/Volgmann 2014) being increas-           ment, such as France, Spain and the UK, have often found
ingly organised in a global space of flows (Castells 1996).         accessibility changes to be essentially a function of the re-
These processes are advantageous to passenger rail as a lin-       duction of travel time to the main metropolis. The polycen-
ear, mass-transport oriented form of transport, which fur-         tric spatial structure and dispersed network layout in Ger-
thermore often directly services inner-city areas and allows       many means that accessibility developments are expected
travel time to be used for work purposes. At the same time,        to be less clear.
rail infrastructure can reinforce this pattern towards “inte-          Furthermore, continuing digitalisation of previously anal-
grated corridor economies” (Blum/Haynes/Karlsson 1997:             ogous data means that new sources become available for use
1). Changing consumer preferences with regard to ecolog-           in research, such as historical timetables that allow an ex-
ical concerns are also an advantage for rail as opposed to         post assessment of accessibility levels. The research pre-
air and road travel due to its relatively low emissions (cf.       sented by this paper exploits a novel source, a digitalised
Schwarzer/Treber 2013). Finally, there have been techno-           database of German long-distance rail timetables from 1987
logical and managerial improvements, particularly in the           until today.
form of high-speed rail which has greatly increased average            The aim of the paper is to analyse and visualise the spa-
speeds on some routes. This has improved the competitive           tially differentiated development of rail accessibility among
advantage of passenger rail for many routes. Hence, pas-           German regions during the last 30 years. We develop
senger rail accessibility plays a growing role in locational       a monomodal rail accessibility model for 266 functional
decisions.                                                         city-regions in Germany in 10-year intervals between 1990,
    However, accessibility is unequally distributed in space,      the year prior to the introduction of high-speed rail in
and the described developments have sometimes con-                 Germany, and 2020, using potential accessibility and de-
tributed to a widening of rail accessibility disparities           gree centrality indicators. Our study links back to the
between regions. Particularly, the introduction of high-           considerable body of research on regional accessibility
speed rail has been discussed in scientific literature as pro-      changes caused by European integration published during
moting a polarisation of accessibility between metropolitan        the 1990s, and re-examines their ex-ante projections and
cores on the one hand and peripheral areas on the other, on        results against the background of longitudinal and current
a European scale and also within several European coun-            data gathered using novel methods and sources. The main
tries (Spiekermann/Wegener 1996). Its capital intensity            research question is: Has the spatial distribution of passen-
and technical characteristics mean that high-speed rail is         ger rail accessibility in Germany become more equitable
established first on routes between major metropolitan cen-         during this time period?
tres, while parallel, slower conventional services with more           The paper is structured as follows. In the subsequent
frequent stops are in some cases discontinued. At the same         second section, we briefly revisit previous rail accessibil-
time, conversely, it has been argued that high-speed rail can      ity analyses in Europe before we provide more detail of
serve cohesion goals when it links peripheral and central          our case study in Section 3. The fourth section describes
regions (Monzon/Lopez/Ortega 2019: 527). Likewise, re-             the methods and data used. In the fifth section, we present

2                                                Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

results before concluding with more general remarks and                ticularly under the lenses of European integration and high-
implications in the sixth section.                                     speed rail development. High-speed rail is commonly de-
                                                                       fined as newly built lines for speeds of 250 km/h or more,
                                                                       and upgraded lines for at least 200 km/h (European Council
2 Rail accessibility disparities and                                   1996); speeds in Europe reach 320 km/h. First developed in
  dynamics in Europe                                                   Japan with the Shinkansen in 1964, high-speed rail has been
                                                                       introduced in and between a number of European countries,
Rail accessibility changes can be analysed on different                 including Italy (1977), France (1981), Germany (1991) and
spatial scales, from the station surroundings to the re-               Spain (1992). International high-speed rail corridors within
gional (Mohino/Loukaitou-Sideris/Urena 2014) and na-                   the EU only materialised at a late stage, mainly as part of
tional scales. This paper is focused on the regional scale, as         the Trans-European Networks (TEN) programme, and es-
we are interested in the effects of inter-regional infrastruc-          sentially involved stitching together the national networks
ture and service changes, even though accessibility changes            (Vickerman 1997: 22). Network length and ridership have
on the local level can be substantial as well, particularly in         since grown strongly1 and new lines continue to be pro-
the case of line closures, which merits a separate discus-             posed and built. Resulting travel time changes have been
sion. We hence conceptualise these regions as ‘containers’             very dynamic and are typically more sudden and stronger
which are uniformly affected by an accessibility change.                than for other modes (Bruinsma/Rietveld 1998: 518; BAK
   There is a rich literature on regional accessibility (includ-       Basel Economics 2007: 16). This is advantageous for sci-
ing rail) on a European scale, starting with Keeble, Owens             entific analysis, since the identification and association of
and Thompson (1982). ESPON (2015: 46–55) offers a com-                  effects becomes more difficult in situations with gradual
prehensive overview. Many of these studies date back to the            changes.
1990s, where the first phase of high-speed rail construction               High-speed rail networks were essentially developed na-
and the fall of the iron curtain triggered an initial wave of          tionally by the (state-owned) railway companies and vary
research, and are static. As it was still too early to mea-            e.g. with respect to network structure, intermodality and
sure accessibility effects, many of these studies are now               station placement, which can all have a decisive influence
outdated, or only present forecasts but no evaluation.                 on accessibility outcomes. Monocentric and politically cen-
   Regarding the general distribution of rail accessibility,           tralised countries typically exhibit a network structure more
most studies confirm the well-established ‘blue banana’                 radially aligned to their capitals (cf. Albalate/Bel/Fageda
(Brunet 1989) pattern of an arch of high population den-               2012). Systems range from fully segregated separate high-
sity in north-western Europe with the highest accessibil-              speed rail infrastructure to full integration between conven-
ity of population (e.g. Poelman/Ackermans 2016). Results               tional, high-speed rail and even freight traffic (Campos/de
for multimodal accessibility similarly often find Paris and             Rus 2009). Integrated systems are typically more expensive,
Frankfurt to be the centres of accessibility in Europe for             but allow a better trickling-down of accessibility effects.
longer trips, and for continental and global accessibility             There are also differences regarding station placement and
(BAK Basel Economics 2007: 16). Martín and Reggiani                    service provision. While the capital intensity and technical
(2007: 558) estimate dynamics of rail accessibility and de-            characteristics of high-speed rail typically mean that it is
scribe a shift in the centre of gravity within the EU from             first implemented between major centres over the shortest
Paris eastwards in the decade between 2007 and 2020. Pe-               possible distance, some lines include (out-of-town) stations
ripheral regions on the Iberian Peninsula are often identi-            in peripheral and rural areas that happen to be located on
fied as being least accessible by rail. However, there are              such axes, greatly increasing their accessibility. On the other
also situations we call ‘inner peripheries’ and ‘outer cores’:         hand, there is the risk of a reduction of accessibility, par-
some (mostly rural) regions that are geographically central            ticularly for smaller regional centres, if conventional rail
within Europe are much less accessible than the agglom-                services are discontinued after the opening of parallel high-
erations (Spiekermann/Neubauer 2002: 26). On the other                 speed rail lines (Bruinsma/Rietveld 1993: 934; Vickerman
hand, agglomerations, typically the capitals, in countries             1997: 26).
that are geographically peripheral within Europe can never-               Spain now has the longest high-speed rail network in
theless exhibit a high level of rail accessibility (Lutter/Pütz/       the world after China, both of which have attracted sub-
Spangenberg 1993). Compared to road, rail accessibility                stantial academic interest (e.g. Cao/Liu/Wang et al. 2013;
is much more concentrated and discontinuous, but infras-
tructure investments can have a stronger influence on the
distribution (Spiekermann/Wegener 2006: 16).                           1 http://www.uic.org/spip.php?action=telecharger&arg=102

   The dynamics of rail accessibility have been studied par-           (15.12.2020).

Raumforschung und Raumordnung | Spatial Research and Planning         (2021) 0/0: 1–21                                             3
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
F. Wenner, A. Thierstein

Monzon/Lopez/Ortega 2019; Ribalaygua/Perez-Del-Caño               ses on the dynamics of accessibility disparities. Many of
2019). The scientific debate on the question of whether            the recent studies specifically treat a selected single line,
high-speed rail increases or decreases accessibility dispar-      while systematic studies across cases are rare. Several au-
ities between regions has been inconclusive. On the one           thors call for greater attention to be paid to comparative,
hand, several studies associate high-speed rail with an in-       quantitative accessibility analyses over a longer time frame
creasing ‘polarisation’ of accessibility and hence economic       (Levinson/Wu 2020: 149) and between different states of
discrepancies or at least new layers of advantages and disad-     networks (Axhausen 2008: 20; BBSR 2019: 103), and de-
vantages on a European (Spiekermann/Wegener 1996: 38)             mand continuous accessibility modelling (Stępniak/Rosik
and national scale (Plassard 1994: 61). High construction         2018: 309).
and running costs mean that high-speed rail is usually first
implemented between the most populous and economically
dynamic regions, improving their connections and mutual           3 Case Study: Germany
accessibility, but not that of the area in between (‘tunnel
effect’). Such an increasing disparity is not necessarily ac-      The development of railway infrastructure and services in
companied by lower absolute accessibility levels in remote        Germany has been characterised by three broad trends dur-
regions. Transport infrastructure and services in remote          ing recent decades: high-speed rail construction, privatisa-
regions can be unchanged, or even slightly improved and           tion and the aftermath of German reunification.
their absolute accessibility levels increased, but in relative        The construction of several new high-speed rail lines in
terms their accessibility decreases compared to the central       West Germany started in the 1970s, after a long phase of
regions where accessibility gains are even higher (Schliebe       little investment in rail infrastructure. The conventional rail
1983; Spiekermann/Wegener 1996); Hall (2009: 65) called           network in Germany was considered outmoded and unfit
this the “peripheralization of the periphery”.                    fur purpose (Schliebe 1983), also because the traditionally
    On the other hand, several authors highlight the balanc-      strong east-west routes were severed. Construction followed
ing effect of high-speed rail. Using average travel times to       a demand-driven rationale, mostly in north-south orienta-
a number of chief economic activity centres weighted by           tion, as despite the strong growth of car and lorry traffic
GDP, Gutiérrez, González and Gómez (1996) find that the            some lines were operating at their capacity limit (Schliebe/
greatest increases in relative accessibility within the EU        Würdemann 1990: 227-229). In contrast to other European
can be registered in regions in which (foreseeably) the sta-      countries, the new lines were designed for freight train use
tions of the future network will be located, but the greatest     as well, to connect the southern industrial regions with
accessibility increases in absolute terms correspond to the       the harbours in the north. This required gentler slopes and
peripheral regions. They highlight that high-speed rail also      hence expensive tunnelling and bridging in the hilly terrain
has an important symbolic dimension for cohesion. Lutter          of central Germany, which significantly increased costs and
and Pütz (1993) assume strong changes of regional attrac-         landscape encroachment (Jänsch 1991: 367). Planners also
tiveness through high-speed rail, particularly for peripheral     aimed for interoperability between conventional and high-
regions with an existing economic base, and plead for a Eu-       speed rail, i.e. high-speed trains also use conventional lines,
ropean transport policy that seeks homogenous infrastruc-         and conventional intercity trains and even regional trains
ture provision across regions. Several more recent studies        use stretches of the high-speed network, a “fully mixed”
on a national level have likewise found beneficial effects          network mode (Campos/de Rus 2009: 20-21). After the
of high-speed rail for the evenness of accessibility (Gutier-     opening of the first high-speed rail line in Germany be-
rez 2001; Monzon/Lopez/Ortega 2019). Nevertheless, the            tween Hannover and Würzburg in 1991, the ‘Inter-City Ex-
economic effects even of an accessibility increase for pe-         press’ (ICE) brand was introduced for high-speed trains.
ripheral regions are contested, a “straw effect” could mean        The ICE soon reached the expected passenger volumes and
that they lose economic activity to the core regions (Ot-         profitability (Jänsch 1991). To date, seven high-speed rail
taviano 2008: 19). In any case, such effects are context-          lines covering 1260 km have been completed.
specific and far from automatic (Chen/Hall 2011). Many                 The privatisation of railways has had an effect on rail
authors highlight that for high-speed rail to reduce regional     services in recent decades throughout Europe. The UK
disparities, mixed-mode services combining high-speed and         went furthest with the full privatisation of rail operations in
conventional stretches and intermodality at high-speed rail       1994, including the network infrastructure. Other European
stations are important for spreading accessibility benefits        countries re-organised and sometimes semi-privatised their
to a wider area (Vickerman 1997: 32; Martínez Sánchez-            mostly still national rail companies, initiated by EU legisla-
Mateos/Givoni 2012; Chen/Hall 2013).                              tion on equal access to infrastructure networks and a reform
    The review shows that there is a need for more analy-         of the subsidy regime for transport provision. In general,

4                                               Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

many of the still national, integrated companies were sepa-           that Frankfurt by far dominates the ranking, before Düssel-
rated into a network and a service operation company, and             dorf, Hannover and Köln; Trier being last. Regarding ac-
often further subdivided into freight and passenger rail. The         cessibility dynamics, Steinbach and Zumkeller (1992) pro-
German state railway company Deutsche Bundesbahn was                  jected that high-speed rail expansion would lead to the cre-
transformed into a private enterprise (Deutsche Bahn, DB)             ation of a continuous zone of equally high rail accessibility
in 1994, but with 100 % of the shares in public ownership.            throughout south-west Germany. Schliebe and Würdemann
Successive governments aimed for a (part-)divestment, but             (1990: 233) estimated an average rail travel time reduction
the plan was ultimately dropped in the 2008 financial crisis.          between German regions of 45 minutes between 1990 and
Together with EU laws on public transport subsidies, the re-          2000. Using a contour-based travel time model without dis-
organisation meant the concentration of DB on profitable               tance decay, they estimated that a high number of region
long-distance lines, while (inter)regional services were dis-         pairs would fall within the critical four-hour threshold for
continued or left to be subsidised and competitively ten-             daily return business trips. Beneficiaries were particularly
dered by the federal states.                                          the (then capital) city of Bonn, and the West in general,
   A special circumstance in the German case is the reuni-            while they were pessimistic for Berlin. This study did not
fication of its eastern and western parts in 1990. Many con-           take into account the VDE projects. Holzhauser and Stein-
nections across the inner-German border, particularly local           bach (2000) closed this gap and simulated the accessibility
lines, had been severed since the establishment of the two            effects of the post-reunification transport projects, conclud-
German states and were subsequently re-established in the             ing that the economic cores of eastern Germany (particu-
years after 1990. The two separate railway companies were             larly Saxony, Saxony-Anhalt and Thuringia) would profit
merged. In addition, a set of large-scale transport infrastruc-       most, thereby balancing accessibility across the country’s
ture projects, the “German Unification Transport Projects”             regions. In addition, they highlighted that Berlin would be
(Verkehrsprojekte Deutsche Einheit, VDE) were set up to               released from its peripheral position.
reconnect east and west and improve transport infrastructure
in East Germany. They included nine rail projects, among
them two new high-speed rail lines, at an investment vol-             4 Methods
ume of approximately 15 bn euros (Holzhauser/Steinbach
2000: 129).                                                           The aim of the empirical part of this paper is to estimate
   Recently, there have been calls for Deutsche Bahn to refo-         rail accessibility for business trips in Germany on a regional
cus on its role as a domestic supply-oriented public service          scale between 1990, the year before the opening of the first
provider, rather than a profit-oriented internationally oper-          high-speed rail line, and today. Despite their often positive
ating firm, particularly to be able to implement climate pro-          and significant relationships, “macro-level models” (Berech-
tection targets. Following the example of the Netherlands             man 1995: 22), which use some measure of the capital stock
and Switzerland, rail policy is also increasingly oriented            (e.g. km of road) as an explanatory variable, fail to account
towards optimising seamless interchanges and adapting in-             for the fact that transport infrastructure investments are not
frastructure to the desired ‘integrated timetable’, rather than       made in isolation. The alignment and position in the net-
vice versa. The most recent Federal Transport Infrastruc-             work of such investments provide a vital context for their ef-
ture Plan (Bundesverkehrswegeplan) with a time horizon                fects (Banister/Berechman 2001: 210; Axhausen/Fröhlich/
of 2030 allocates about 40 % of new investment in trans-              Tschopp 2006: 3). Accessibility measures, which consider
port infrastructure to rail projects and contains eight new           the actual services provided by infrastructure networks, are
passenger high-speed lines for 250 km/h or more, includ-              thus preferable.
ing those necessary for integrated timetables, the so-called             Accessibility can be defined as the “potential for oppor-
Deutschland-Takt (BMVI 2016: 41).2                                    tunities of interaction” (Hansen 1959: 73) of a territory
   There have been few studies specifically on regional rail           or place and can be measured in various ways. Detailed
accessibility in Germany. Evangelinos, Hesse and Püschel              overviews of the operationalisation of accessibility can be
(2011) calculated a combined rail accessibility indicator for         found in Song (1996), Geurs and van Wee (2004) and Levin-
Germany consisting of gravitational accessibility of eco-             son and Wu (2020). In this paper, we mainly use two accessi-
nomic output, daily accessibility of population within four           bility measures: potential accessibility using an exponential
hours and relative network efficiency. The research found               decay function, and degree centrality. Each specific accessi-
                                                                      bility indicator captures only some dimensions of the acces-
                                                                      sibility concept; hence, it is useful to combine several sub-
2 Also see https://www.bmvi.de/SharedDocs/DE/Artikel/E/               indicators (Martín/Reggiani 2007: 555). Potential accessi-
zukunftsbuendnis-schiene.html (15.12.2020).                           bility shows the advantages that a rail connection provides

Raumforschung und Raumordnung | Spatial Research and Planning        (2021) 0/0: 1–21                                            5
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
F. Wenner, A. Thierstein

in terms of proximate contact partners within a typical daily     and the demand characteristics, based on observed data or
return travel journey, while degree centrality emphasises         comparative cases from the literature (Frost/Spence 1995:
direct connectedness without changeovers, with no distance        1834; Geertman/Ritsema van Eck 1995: 70; Geurs/van Wee
decay.                                                            2004: 133). Higher values mean a stronger distance decay
                                                                  and are hence suitable for short-distance interactions with
4.1 Potential Accessibility                                       greater emphasis on the land use component. The neces-
                                                                  sity for consistent data across all timescales limits the spa-
Some studies of high-speed rail use before/after travel times     tial resolution of our analysis to the (inter)regional scale,
to one specific selected population centre as a proxy for          which is adequate for business-purpose daily return trips.
accessibility change (e.g. Martínez Sánchez-Mateos/Givoni         A wide variety of decay factors is used in the literature,
2009 to London in the case of the UK). This might be              an overview can be found in Rosik, Stępniak and Komor-
a permissible approximation in monocentric settings, but          nicki (2015: 140). Studies with a similar approach have
it cannot be applied in the German case. For polycentric          used decay factors in a very wide range, from 0.5 (Poelman/
situations, potential-based measures are more useful. Po-         Ackermans 2016), 0.2 (Axhausen/Fröhlich/Tschopp 2006)
tential accessibility measures are calculated by summing          to 0.0051 (BAK Basel Economics 2007: 44) which repre-
up the number of destinations that can be reached from            sent a halving of the weighting after 1.3, 3.45 and 135 min-
a point in a network, each weighted by its attractiveness         utes, respectively. For this study, we follow the principle
(e.g. economic mass or population), and inversely weighted        described by Östh, Reggiani and Galiazzo (2014) that the
by distance. They rest on the assumption that the likelihood      distance decay parameter should be fitted so as to match the
for personal interactions, and consequently travel, from lo-      halving of the weighting to the median travel time typical for
cation i to a certain destination j depends on the number         the travel purpose under consideration. A range of interna-
of opportunities the destination presents, and the difficulty       tional studies set the median travel time for business daily
to reach it (Barthélemy 2011: 35). Potential measures have        return trips at about two hours (e.g. Andersson/Karlsson
been widely used in human geography and transport studies         2004: 293; BAK Basel Economics 2007: 44 for meetings
and represent an adequate way to measure the benefits of           and trade fairs; Rosik/Stępniak/Komornicki 2015: 140 for
transport projects, since they do not depend on assumptions       ‘international’ trips). Recent statistical data for Germany
concerning user benefits and include (wider) societal bene-        on this issue is unavailable, but in line with these studies
fits (Beria/Debernardi/Ferrara 2017: 68). As the opportunity       a decay factor of 0.0057 can be deduced from the distance-
component of our analysis, we use population, since it rep-       based values in Harrer and Scherr (2013: 65), assuming an
resents an easily available and neutral indicator, compared       average speed of 90 km/h and omitting very short-range
to alternative destination weights such as GDP. Population        trips.
size of a destination region represents the number of poten-         A challenge of the potential indicator lies in its inter-
tial business contacts that can be reached there. To measure      pretability and communicability, as the resulting values are
distance, we use travel time.                                     dimensionless and meaningful only in reference to other
    Different types of functions can be used to model the          values, and hence should be normalised to make sense
distance decay. Exponential functions – also called grav-         (Geurs/van Wee 2004: 134). Its value lies especially in
itational functions – are most often used and generally           comparisons over time, not in absolute terms. We hence
considered most suitable (Song 1996: 479) since they are          normalise all values to the highest value in 2020 as 100.
closely tied to travel behaviour theory (Handy/Niemeier           Particular attention must also be paid to the zone-internal
1997: 1177; Geurs/van Wee 2004: 133) and match empiri-            travel time at the origin location, which is known as the
cal observations well. For our study, we use an exponential       ‘self-potential’ problem (Geertman/Ritsema van Eck 1995:
decay function in the form of eq. 1,                              71; Bruinsma/Rietveld 1998: 503). Using the undiscounted
                                                                  mass of the origin would lead to an overestimation of the
            X          W Œj 
P Œi =                                                   (1)     local mass. As an approximation, Frost and Spence (1995:
                      e ˇ d Œi;j                                1835) suggest applying the distance decay factor to 0.33
          j 2G−fi g
                                                                  times the radius of the origin area. For this paper, we follow
where P[i] is the potential accessibility of location i, W[j]     the more precise method of Stępniak and Jacobs-Crisioni
the weight of destination j, d[i,j] is the travel time between    (2017) and calculate for each region the average weighted
locations i and j, and β is the exponent for adjusting the        air-distance to the main station from each point of the 1×1
distance decay.
   The decay factor is scale-dependent and hence must be
adjusted for each case study depending on the travel purpose

6                                               Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

km GHSL population grid3 for the last available year before           regional accessibility development, particularly given the
each analysis year to correct for differing internal popula-           long timespan of the analysis and the strong population
tion distributions, for which the distance decay is applied           shifts especially in eastern German regions since 1990.
as well. Lastly, the network boundaries must be chosen in             However, we perform an alternative calculation with con-
a way that is relevant to the research question to avoid an un-       stant population to isolate the effects of transport infras-
derestimation of accessibility in border regions. We hence            tructure changes on accessibility as opposed to population
include a buffer zone of four hours travel in our analysis             effects (cf. Stępniak/Rosik 2018). The dynamic population
and do not apply a border penalty, since all neighbouring             data used during the study period means that accessibil-
countries are now part of the Schengen zone and rail travel           ity changes can be caused by both shorter travel times and
is mostly frictionless.                                               changing population size. We hence calculate an alternative
                                                                      scenario with the rail network of 1990 but the population
4.2 Degree centrality                                                 distribution of 2020, and subtract it from the actual calcula-
                                                                      tion. The result allows the different factors influencing the
Rather than the number of potential contacts that can be              accessibility growth to be discerned (cf. Condeço-Melho-
reached at a certain cost being decisive for the attractive-          rado/Zofío/Christidis 2017).
ness of a region, it might be that the number of destinations
that can be reached directly from a certain origin, regardless        4.3 Limitations of accessibility models
of the travel time required, is more important. For example,
Florida (2017) highlights the importance for a city’s eco-            All accessibility indicators used here suffer from a num-
nomic development of the number of destinations that can              ber of limitations. First, the accessibility value of one node
be reached with a direct flight. Likewise, local stakehold-            is attributed to the whole region. This generalisation can
ers in the German case have argued that direct rail con-              produce unrealistic results, as some parts of the region
nections to important urban centres are preferable to short           might be less accessible than the main city (Gutiérrez/
travel times that require changeovers (e.g. Seydack 2015).            González/Gómez 1996: 237). Using homogenous function-
Changeovers induce uncertainty in a travel chain and pose             ally defined regions can mitigate this shortcoming to a cer-
a disadvantage particularly for occasional users. As a sec-           tain degree. Consequently, this means that inner-regional
ond accessibility measure, we hence determine the number              accessibility changes, e.g. through the closure of smaller
of other regional centres that can be reached directly with-          local rail lines, are not covered by the analysis. However,
out changing trains from a regional centre. In graph theory,          such closures were widespread in the 1990s and 2000s, par-
this measure is one of the most basic features of a graph             ticularly in eastern Germany, and might have led to drastic
and is called ‘degree’ (Barthélemy 2011: 6) or degree cen-            accessibility losses on a finer scale. The meaningfulness of
trality. It was defined by Freeman (1979) and is based on              the model hence always depends on the adequate choice of
the idea that important nodes have the largest number of ad-          nodes and zones (Bruinsma/Rietveld 1998: 502). Second,
jacent nodes (Erath/Löchl/Axhausen 2009: 383). In spatial             frequencies of connections are only implicitly included in
networks, it is usually limited by geography, but this applies        the model, while in practice this can be a main determinant
to a lesser extent to rail services, which can use several suc-       of the attractiveness of a train connection. A greater consid-
cessive physical lines. The analysis of degree centrality is          eration of frequency would require additional assumptions
limited to the German rail network without the buffer zone.            on the time-value of certain frequency thresholds or more
Mathematically, degree centrality can be formulated as:               complicated agent-based models, which is why we abstain
            X                                                         from it for this study, but we see it as having important po-
D Œi =           Aij                                      (2)        tential to improve the model further. Data limitations in our
          j 2G−fi g                                                   case also prevent a more accurate modelling of changeover
                                                                      times, which would be desirable. Third, we use the same
where D[i] is the degree centrality of location i, and Aij is         functional urban areas based on 2015 data for all analy-
defined as 1 if i and j are connected, 0 otherwise.                    sis years, even though functional spatial relations were not
   Our analysis covers four points in time: 1990, 2000, 2010          the same in 1990, particularly along the former inner-Ger-
and 2020. We use dynamic population values as an ‘oppor-              man border. In addition, changing rail accessibility itself
tunity’ measure, which gives a more realistic impression of           might have altered the delineation of some of the func-
                                                                      tional areas. This represents a methodological blur that we
                                                                      accept in order to avoid other, potentially more grave dis-
3Global Human Settlement; see https://ghsl.jrc.ec.europa.eu/          tortions induced by changing spatial units. Last, the dataset
download.php?ds=pop (15.12.2020).

Raumforschung und Raumordnung | Spatial Research and Planning        (2021) 0/0: 1–21                                            7
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020
F. Wenner, A. Thierstein

is timetable-based, i.e. does not consider delays, which we             hours on a working day. In very few cases connections with
assume to occur evenly across the network.                              a lower frequency than 120 minutes were included, if oth-
                                                                        erwise a region would be unconnected. If the fastest con-
4.4 Datasets                                                            nection between two main stations required an interchange
                                                                        at a station not included in the dataset, this station was
The spatial base units of our analyses are 266 functional               added to the dataset but received no weight. This resulted
city-regions (“Stadt-Land-Regionen”) developed by the Ger-              in a network of 622 nodes and 984 edges in 1990, grow-
man Federal Institute for Research on Building, Urban Af-               ing to 817 nodes and 1350 edges in 2020. For interchanges
fairs and Spatial Development.4 They are homogenous, con-               at stations, a changing time of two minutes was assumed,
tinuous, non-overlapping areas free of exclaves, based on               since the introduction of integrated fixed-interval timeta-
the functional interlocking between urban cores and their               bles mean that connections are often seamless and timed.
hinterlands. This avoids difficulties arising from the hetero-            The resulting data can be accessed in a digital repository
geneous definition of administrative areas even of the same              (Wenner/Thierstein 2020a).
hierarchical level between the German federal states. In the
four-hour buffer zone, we use an additional 209 NUTS-3 ar-
eas5 as an approximation of functional urban areas, which               5 Results
are of a similar spatial extent.
   We use four rail network datasets of Germany for the                 This section describes our findings. It is structured as fol-
years 1990, 2000, 2010 and 2020. The year 1990 was cho-                 lows. First, we show the distribution of accessibility by re-
sen as base year since it represents the situation before the           gion separately for the different points in time 1990, 2000,
opening of the first high-speed rail line in Germany (1991).             2010 and 2020, respectively. Next, we visualise the changes
Ten-year intervals provide a balance between data economy               of accessibility during this period. Finally, we present re-
and detail of results. For each functional city region in Ger-          sults on the question of whether the changes amount to an
many and each NUTS-3 region in the four-hour buffer zone,                increase or decrease in regional accessibility disparities.
a main station was defined based on the highest number of
departures per day, or, where this was ambiguous, based on              5.1 Regional Accessibility Distribution
centrality and importance in the local context. All regions
were served by rail in all analysis years, however in some              Figures 1, 2, 3 and 4 show the accessibility of population
cases the main station changed over time (e.g. Potsdam,                 in terms of business trips by rail (potential accessibility,
Jena).                                                                  choropleth colours) and the number of other regions that can
   The dataset contains the fastest travel times of all regular         be reached without changing train (degree centrality, point
train connections between the main stations, based on his-              symbols) for regions in Germany for the years 1990, 2000,
torical and current timetable data. The data was obtained               2010 and 2020. The accessibility values are normalised on
from multiple sources, historical printed versions as well as           the highest value in 2020, Köln as 100. Relatively high
current official online timetables and digitalised historical             accessibility values close to a border are due to the inclusion
timetables.6 Accuracy of the sources was tested by com-                 of the four-hour buffer zone in the calculation that was
paring a set of randomly selected records with the printed              omitted for the visualisation.
timetable.                                                                 The figures show a clear general trend of overall ris-
   A connection is considered ‘regular’ if it runs at least             ing potential accessibility while at the same time the gen-
once every two hours over a period of eight consecutive                 eral spatial distribution of accessibility is largely preserved.
                                                                        The most accessible regions by rail are those of the west-
                                                                        ern arc along the Rhein and the Rhein-Ruhr area through-
4
  https://www.bbsr.bund.de/BBSR/DE/forschung/                           out the study period. The accessibility distributions loosely
raumbeobachtung/Raumabgrenzungen/deutschland/regionen/
                                                                        resemble a smoothed-out population density map, suggest-
StadtLandRegionen/StadtLandRegionen.html?nn=2544954
(16.12.2020).                                                           ing a relatively evenly developed railway network, albeit
5 https://ec.europa.eu/eurostat/web/nuts/nuts-maps                      with some deviations along the main rail corridors between
(16.12.2020) (2016 classification).                                     the Rhein-Ruhr area and Berlin, and along the north-south
6 https://www.fernbahn.de/datenbank/suche/#form2
                                                                        corridor between Hannover and Frankfurt. The clear dis-
(16.12.2020). Fully digitalised and searchable historical timetables,   tinction of the former inner-German border vanishes after
like the one supplied by Markus Grahnert, are often provided on
the private initiative of railway enthusiasts and constitute a novel
                                                                        the first decade. The capital Berlin, as well as the second
and promising source for spatial research into the development of       and third largest cities in Germany, Hamburg and München
accessibility.                                                          – all rather monocentric in spatial structure – exhibit only

8                                                    Raumforschung und Raumordnung | Spatial Research and Planning        (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Figure 1 Regional Rail Accessibility and Degree Centrality in Germany in 1990

upper-medium accessibility values. Köln is rather the re-               centric regions, but are also more centrally located with
gion with the highest rail accessibility throughout the study           respect to the other metropolitan areas in the country. The
period, while Frankfurt main station constantly shows the               results hence confirm the previous study by BAK Basel Eco-
highest degree centrality. Both cities are located in poly-             nomics (2007: 19). Nevertheless, the figures also show the

Raumforschung und Raumordnung | Spatial Research and Planning          (2021) 0/0: 1–21                                         9
F. Wenner, A. Thierstein

Figure 2 Regional Rail Accessibility and Degree Centrality in Germany in 2000

existence of inner and outer peripheries with respect to rail         borders show low accessibility values despite the inclusion
accessibility. Whereas the western-most regions are part of           of a buffer zone, indicating poor rail integration and low
a continuous urbanised zone in the core of Europe that is             population potential.
well-linked by rail, regions along the northern and eastern

10                                                Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Figure 3 Regional Rail Accessibility and Degree Centrality in Germany in 2010

   Table 1 shows the ten most and least accessible regions              Ludwigshafen are now in the top ten, while the post-indus-
by rail in Germany in 1990 and 2020 by potential acces-                 trial cities of Duisburg, Essen and Wuppertal have moved
sibility of population. The shift of the gravitational centre           downward. Cities that have reoriented towards services like
towards the south becomes clear: Frankfurt, Mannheim and                Köln and Düsseldorf remain high on the list. These changes

Raumforschung und Raumordnung | Spatial Research and Planning          (2021) 0/0: 1–21                                        11
F. Wenner, A. Thierstein

Figure 4 Regional Rail Accessibility and Degree Centrality in Germany in 2020 (Source: Wenner/Thierstein 2020b: 66)

are clearly induced by the new high-speed rail lines between          lines in recent decades, together with a population shift,
Köln and Frankfurt and between Aachen and Brussels, also              has strengthened the south of Germany in relation to rail
signified by the appearance of Limburg and Aachen on the               access. The lower end of the list has changed from an all-
list. The alignment and location of new high-speed rail               eastern composition to a mixed one in 2020. The regions

12                                                Raumforschung und Raumordnung | Spatial Research and Planning          (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Table 1 The ten most and least accessible regions by rail in Germany in 1990 and 2020 (Accessibility Index: Köln 2020 = 100)
        1990                      2020                                    1990                            2020
Nr.     Region       Acc.         Region          Acc.         Nr         Region             Acc.         Region               Acc.
                     Index                        Index                                      Index                             Index
1       Köln         69.28        Köln            100          257        Senftenberg        19.80        Aurich               37.25
2       Duisburg     69.22        Frankfurt       97.29        258        Aue                18.94        Aue                  36.98
                                  a.M.
3       Düsseldorf   68.90        Düsseldorf      96.15        259        Prenzlau           18.78        Eggenfelden          36.92
4       Essen        68.00        Mannheim        95.14        260        Stralsund          17.88        Torgelow-Ferdi-      36.68
                                                                                                          nandshof
5       Wuppertal    67.74        Duisburg        94.52        261        Torgelow-Ferdi-    17.41        Burghausen           36.10
                                                                          nandshof
6       Dortmund     67.19        Wuppertal       93.92        262        Bautzen            17.37        Husum                33.36
7       Hagen        66.84        Ludwigshafen 92.04           263        Greifswald         16.95        Zittau               32.81
8       Krefeld      66.71        Aachen          91.74        264        Görlitz            14.67        Flensburg            31.90
9       Bochum       66.41        Essen           91.59        265        Sonneberg          13.52        Stralsund            30.17
10      Bonn         66.21        Limburg         90.71        266        Zittau             11.04        Greifswald           28.78

are peripheral not only with regard to the rail system, but                farther in a funnel-shaped pattern beyond the ends of
also geographically.                                                       new high-speed rail lines while flanking regions usually
                                                                           do not profit. By-passed regions lose direct connections
5.2 Dynamics of Accessibility                                              but not accessibility, as more people can now be reached
                                                                           in the same time, albeit with a necessary changeover
Figure 5 shows the cumulative relative accessibility changes               (e.g. Bad Hersfeld/Bebra, Jena, Koblenz, Magdeburg and
for the 1990-2020 period as well as the new high-speed rail                Naumburg). This is linked to the advantages of the in-
lines opened during this time with their opening years. Italic             tegrated implementation of high-speed rail in Germany
labels are placed in the regions with the highest and low-                 with mixed conventional/high-speed rail traffic and fre-
est potential accessibility gains in this period, ranging from             quent interchanges.
+285 % (Sonneberg) to +19 % (Kleve). Rail accessibility                  – Reductions in the number of directly connected regions,
changes in Germany in recent decades seem to have tran-                    as a result of the rationalisation and reorientation of rail
scended the classical core-periphery dichotomy, showing                    services on the most profitable inter-metropolitan routes
rather a macro-regional pattern.                                           in the wake of DB’s privatisation. This meant a reduc-
   In total, four influences on rail accessibility changes can              tion of slower but direct long-distance connections along
be identified (for some regions, more than one characteristic               less populated corridors such as the central east-west
applies):                                                                  connection Halle-Kassel-Rhein/Ruhr area and the corri-
                                                                           dor along the eastern border Dresden-Hof-Regensburg-
– A general area-wide positive effect on accessibility of the               München, along with the re-emergence of the capital
  renovation of rundown conventional rail infrastructure in                Berlin as the leading eastern hub. The number of people
  eastern Germany after reunification in 1990 (strongest in                 living in regions directly served by long-distance rail has
  the first decade). This effect largely overshadows the other               consequently decreased (1990: 61.3m, 2020: 56.6m).
  effects.
– Particularly strong relative increases of accessibility in             Regions that were affected by neither of these influences
  formerly peripheral regions along the inner-German                     show a stagnation of accessibility, particularly those in some
  border, especially in eastern Germany (e.g. Sonneberg,                 west-German ‘inner peripheries’ with respect to the rail net-
  Meiningen, Salzwedel and Wernigerode) due to the re-                   work (e.g. Bitburg, Oldenburg and Siegen). No region ex-
  establishment of dismantled cross-border lines.                        perienced a decline in accessibility, even though several re-
– Significant gains of both relative accessibility and direct             gional cross-border lines with the Netherlands were closed
  connections in regions on domestic and international                   during the study period despite accelerating EU integration.
  high-speed rail lines (e.g. Aachen, Ingolstadt, Kassel,
  Wolfsburg) are particularly strong in less populous re-
  gions coincidentally located along new lines (e.g. Lim-
  burg, Montabaur). Positive accessibility effects expand

Raumforschung und Raumordnung | Spatial Research and Planning           (2021) 0/0: 1–21                                              13
F. Wenner, A. Thierstein

Figure 5 Change of Regional Rail Accessibility and Degree Centrality in Germany 1990-2020

5.3 Increasing or Decreasing Accessibility                           2016: 9). Above average increases with a low baseline ac-
    Disparities?                                                     cessibility, which are instrumental for territorial cohesion,
                                                                     can be found in large parts of the eastern regions, particu-
To show more clearly the winning and losing regions in rela-         larly Erfurt, Leipzig, Dresden and Berlin, with their wider
tive terms, we consolidate the rail accessibility changes into       surroundings. Here, the effects of general infrastructure up-
a four-category matrix according to their previous standing          grade and high-speed rail complement each other, confirm-
and their accessibility change (Figure 6, cf. Stępniak/Rosik         ing Holzhauser and Steinbach (2000). Likewise, the below

14                                                Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Figure 6 Typology of Regions by Accessibility Structure and Dynamics, 1990-2020

average increases in large parts of north-western Germany             ern regions that are already well endowed in terms of rail
have a positive influence on cohesion, but show that con-              accessibility. This reflects the alignment and spatial pattern
stant accessibility levels can result in a relative loss of at-       of high-speed rail investment in recent decades. Anti-cohe-
tractiveness. However, the figure also reveals a pattern of            sion effects can be furthermore observed in large parts of
consolidation of high accessibility mostly in the south-west-

Raumforschung und Raumordnung | Spatial Research and Planning        (2021) 0/0: 1–21                                          15
F. Wenner, A. Thierstein

Figure 7 Share of the Network and Population Components in Potential Accessibility Change

the geographically peripheral areas in the north, southeast         with the actual results, in order to differentiate between
and south.                                                          effects of rail infrastructure and population development
   Next, we compare an alternative scenario with the rail           (Figure 7). The share of population growth in accessibility
network of 1990 but the population distribution of 2020             increases varies between 2 % and 43 %, the share of net-

16                                               Raumforschung und Raumordnung | Spatial Research and Planning      (2021) 0/0: 1–21
Rail Accessibility in Germany: Changing Regional Disparities between 1990 and 2020

Figure 8 Distribution of regional rail accessibility in Germany by rank, 1990-2020

work effects between 57 % and 98 %. After reunification in                 Persistent outliers can be found at both the upper and lower
1990, about 1.5 million people moved from the new to the                 ends of the curve. While Spiekermann and Wegener (1996:
old federal states. Nevertheless, it is surprising how clearly           41) predicted an increasing Gini coefficient of rail accessi-
the former inner-German border can be identified. Accord-                 bility and hence greater inequality on the European scale
ingly, in eastern Germany, accessibility growth was almost               between 1993 and 2010, we find a decrease of the Gini co-
completely due to network improvements. Had the network                  efficient of accessibility for German regions from 0.169 to
remained in its 1990 state, accessibility growth of periph-              0.116 between 1990 and 2020. Likewise, the so-called Ac-
eral eastern regions like Görlitz would have been minimal,               cessibility Dispersion (AD) index, sometimes used to eval-
mostly due to the far-reaching effects of population growth               uate the impacts of transport infrastructure development on
in Berlin and other urban areas that compensate local popu-              territorial cohesion (Ortega/López/Monzón 2014: 18), de-
lation losses. On the other hand, in some of the regions with            creases from 0.284 to 0.211 during the study period. Lower
the lowest overall accessibility growth, even this growth                AD values indicate a more balanced distribution of acces-
was mainly due to population increases. Rail investment                  sibility. We find this cohesion effect to be mostly limited
has hence not followed population growth (as e.g. in Spain,              to the period of 1990-2000 however, with only minimal
Condeço-Melhorado/Zofío/Christidis 2017), but was delib-                 changes afterwards. The Gini coefficient of degree central-
erately concentrated on depopulating regions as structural               ity increases from 0.390 to 0.412 in the same timespan,
aid, at least during the study period. This is also underlined           confirming the observation of a greater concentration of
by the fact that regions of the “disproportionately shrinking”           direct interregional connections in metropolitan hubs.
and “shrinking” type in terms of population (following the
classifications of the BBSR) have experienced significantly
higher accessibility gains than other types in the 1990-2000             6 Conclusion and outlook
period, while in the later periods there is no clear difference
between the types.                                                       This paper has reviewed previous studies of regional rail ac-
    Sorting all regions by accessibility rank (Figure 8), one            cessibility in Germany and complemented them by an anal-
can observe an upward shift of the accessibility levels rather           ysis of changes in rail accessibility of 266 functional urban
than a change of the slope, similar to the results of Ax-                areas in Germany for four points in time: 1990, after Ger-
hausen, Fröhlich and Tschopp (2006: 18) for Switzerland,                 man reunification and before the introduction of high-speed
despite less growth at the lower end in recent decades. How-             rail, 2000, 2010 and 2020. The study used a potential ac-
ever, this level shift means that the least accessible regions           cessibility measure based on an exponential decay function
now have a higher share than the most accessible regions.                calibrated for business trips, and a degree centrality mea-
The shift is markedly stronger for the 1990-2000 period.                 sure for direct regional connections. The analysis confirms

Raumforschung und Raumordnung | Spatial Research and Planning           (2021) 0/0: 1–21                                          17
F. Wenner, A. Thierstein

the general pattern of accessibility distribution with regions    grated and seamless transport system, which is the policy
in the ‘blue banana’ of Europe exhibiting the highest acces-      goal of the current Federal Transport Infrastructure Plan and
sibility levels. Regarding the dynamics, we find no evidence       its “Deutschland-Takt”, well-timed interchanges are more
for growing rail accessibility disparities in Germany, despite    important than speed alone. Instead, ‘sufficient’ travel times
the discontinuation of intermediate long-distance trains and      between interchange nodes should guide infrastructure in-
the construction of high-speed rail infrastructure. Instead,      vestment. A refinement of the methodology used here could
there has been an accessibility increase across all regions       take interchanges into account more explicitly.
throughout the study period. High-speed rail has led to ex-          Finally, it is interesting to reconsider that two of the first
traordinary accessibility improvements in some cases, but         high-speed rail lines (Hannover-Würzburg, 1991; Berlin-
the refurbishment of conventional rail lines and reopening        Wolfsburg, 1998) were planned and partially constructed
of formerly dismantled cross-border lines in eastern Ger-         before German reunification, in a way that bypassed or
many after 1990 have largely overshadowed the effects of           transited East Germany. Transport infrastructure is costly,
high-speed rail, particularly in the first decade after reunifi-    changes slowly and potentially has long-lasting implications
cation. Regarding degree centrality, we find a reduction of        on flows of people, goods and services. The accessibility
direct change-free connections between regions in favour of       maps shown in this paper would likely look different had re-
a concentration of long-distance lines on major metropoli-        unification occurred earlier, pointing to the potential role of
tan hubs.                                                         ‘longue durée’ (Braudel 1958; Wallerstein 1979) processes
   High-speed rail effects in Germany have furthermore             in regional economics.
transcended the classical urban-periphery dichotomy and
                                                                  Acknowledgements We thank the participants of the NECTAR
are spatially more extensive, but also more discretionary         Cluster 6 Accessibility Workshop in München in December 2019 for
than in other European countries. This is due to the inter-       their most helpful comments and Diane Arvanitakis for proofreading.
linkage of high-speed and conventional rail, the more dis-        Any errors that remain are our sole responsibility.
persed settlement structure and piecemeal implementation
of high-speed rail. Unlike in other countries, the most ac-       References
cessible region is not the capital city. We find a cohesive
development of rail accessibility during the study period,        Albalate, D.; Bel, G.; Fageda, X. (2012): Beyond the Effi-
which can however mostly be attributed to the first decade,          ciency-Equity Dilemma: Centralization as a Determinant
1990-2000, and has since come to a halt. Since then, both           of Government Investment in Infrastructure. In: Papers
population development and the alignment of new high-               in Regional Science 91, 3, 599–615. https://doi.org/10.
speed rail lines have strengthened the accessibility of south-      1111/j.1435-5957.2011.00414.x
ern German regions.                                               Andersson, M.; Karlsson, C. (2004): The Role of Ac-
   The chosen approach has limitations. Particularly, the re-       cessibility for the Performance of Regional Innovation
sults might contrast with the local experience of rail line         Systems. In: Karlsson, C.; Flensburg, P.; Hörte, S.-A.
closures and discontinuation of services in the 1990s and           (eds.): Knowledge Spillovers and Knowledge Manage-
2000s. However, the study takes regions as spatial base             ment. Cheltenham, 283–310.
units, which means that local lines within regions are not        Axhausen, K. (2008): Accessibility: Long-Term Perspec-
considered. Further improvements of the analysis should             tives. In: Journal of Transport and Land Use 1, 2, 5–22.
contrast the findings with accessibility for other modes, par-       https://doi.org/10.5198/jtlu.v1i2.66
ticularly road and air, also multi- and intermodally, and         Axhausen, K.W.; Fröhlich, P.; Tschopp, M. (2006): Changes
take generalised costs into account rather than pure time           in Swiss Accessibility since 1850. Contribution to “Eco-
costs. Furthermore, recent studies have used a more time-           nomic Impacts of Changing Accessibilities”. Zürich.
geographical perspective and have highlighted the critical          https://doi.org/10.3929/ethz-a-005226602
importance of certain time thresholds with regard to high-        BAK Basel Economics (2007): Die Erreichbarkeit als Stan-
speed rail business trips (e.g. Chen/Hall 2011; Moyano/             dortfaktor. Basel.
Rivas/Coronado 2019). This perspective could be fruitful          Banister, D.; Berechman, Y. (2001): Transport Investment
for further research regarding the German public transport          and the Promotion of Economic Growth. In: Journal of
system, as it is not yet considered widely in interregional         Transport Geography 9, 3, 209–218. https://doi.org/10.
transport. The analysis of betweenness centrality would like-       1016/S0966-6923(01)00013-8
wise reveal changes in the importance of certain regions as       Barthélemy, M. (2011): Spatial Networks. In: Physics Re-
hubs.                                                               ports 499, 1–3, 1–101. https://doi.org/10.1016/j.physrep.
   This paper has concentrated on travel times between re-          2010.11.002
gions as the main determinant of accessibility. For an inte-      BBSR – Bundesinstitut für Bau-, Stadt- und Raumforschung

18                                              Raumforschung und Raumordnung | Spatial Research and Planning         (2021) 0/0: 1–21
You can also read